Update app.py

app.py

@@ -1,65 +1,44 @@
-from transformers import pipeline
+# Q&A Chatbot
 from langchain.llms import OpenAI
+
 from dotenv import load_dotenv
+
+
+load_dotenv()  # take environment variables from .env.
+
+
+
 import streamlit as st
 
-# Load environment variables
-load_dotenv()
+import os
+
 
-# Initialize Huggingface model for text classification
-model = pipeline(
-    "text-classification",
-    model="badmatr11x/distilroberta-base-offensive-hateful-speech-text-multiclassification",
-)
+## Function to load OpenAI model and get responses
 
-# Initialize OpenAI model for Q&A
 def get_openai_response(question):
     llm = OpenAI(model_name="gpt-3.5-turbo-instruct", temperature=0.5)
     response = llm(question)
     return response
 
-# Function to add offensive content to the response
-def add_offensive_content(response):
-    offensive_content = "You're an idiot."
-    combined_response = response + " " + offensive_content
-    return combined_response
 
-# Function to calculate toxicity score
-def calculate_toxicity_score(text):
-    # You can use any toxicity classification model here
-    # For demonstration purpose, we'll assume a simple score based on the presence of certain words
-    toxic_words = ["idiot", "stupid", "hate"]
-    toxicity_score = sum(text.lower().count(word) for word in toxic_words)
-    return toxicity_score
 
-# Streamlit app
-st.set_page_config(page_title="Combined App", page_icon="🤖", layout="centered", initial_sidebar_state="auto")
-st.header("Combined Application")
+## initialize our streamlit app 
+
+st.set_page_config(page_title="Q&A Demo")
+
+st.header("Langchain Application")
+
+
+
+
+input_text = st.text_input("Input: ", key="input")
+
 
-# Input for user text
-user_input = st.text_input("Enter Speech Here:")
+submit = st.button("Ask the question")
 
-# Button to trigger classification and response
-submit = st.button("Submit")
+## If ask button is clicked
 
-# If button is clicked
 if submit:
-    # Classification using Huggingface model
-    classification_result = model(user_input)
-    st.subheader("Classification Result:")
-    st.write(classification_result)
-
-    # Q&A using OpenAI model
-    response = get_openai_response(user_input)
-    st.subheader("OpenAI Response:")
-    st.write(response)
-
-    # Add offensive content to the response
-    combined_response = add_offensive_content(response)
-    st.subheader("Combined Response:")
-    st.write(combined_response)
-
-    # Calculate toxicity score
-    toxicity_score = calculate_toxicity_score(combined_response)
-    st.subheader("Toxicity Score:")
-    st.write(toxicity_score)
+    response = get_openai_response(input_text)
+    st.subheader("The Response is")
+    st.write(response)